Chatbot User Interface for Customer Relationship Management using NLP models

NLP is the most researched field. Speech-totext conversions, fake-news detection, and text summarization are the hot topics of NLP. ChatBot User Interface(UI) using NLP, allows machines to understand customers better. The aim was to use different NLP and machine learning techniques and to add ChatBot UI to guide customers or clients through the CRM software and help them whenever they get stuck. Different approaches, libraries, and algorithms like 'RASA', python's 'Chatterbot', 'Cosine similarity', and Google's embedder were used to train the model and then later compared to see which gave the best results. After that, during the deployment other 2 approaches were tried, one was fetching questions from the database and then training the model, the other was to maintain a local text document and train the model from that. The advantages and disadvantages of each approach, plus challenges and better methods for deployment is also discussed.